We have released two new preprints presenting advanced machine learning applications for particle reconstruction:
“ParticleTransformer is all you need for reconstructing hadronic tau leptons” (arXiv:2606.18460): Presents the first fully machine-learned hadronic tau reconstruction approach for the future FCC-ee collider using the CLD detector setup, demonstrating sub-percent transverse momentum resolution.
“Machine-learned particle flow as a foundation model for collider physics” (arXiv:2606.14373): Establishes ML-based particle flow (MLPF) as a foundation model by showing that its per-particle latent representations encode essential physics information that significantly improves downstream tasks like jet flavor tagging and energy regression.